Wolfgang Schwarz

>"If whether I believe P or Q depends on far away events, how can I know I believe P rather than Q if I don't know about these far away events?"<

Well, I think, you have three possibilities:

First to ignore the problem all the way. Call this the Burgean strategy. In order to take that strategy, you admit that we cannot discriminate P from Q. With a second step you point out that our second order thoughts (which are in accordance with the following schema:
I think that I think that p)
are always true. We simply cannot be wrong with regard to our relevant first order thought. Therefore: When we think, that we think that p, then we know that we think that p. ...and be happy.

Second possibility: Recognize the problem and claim that we don't know that we think that p, when we think that we think that p.Call this the Boghossian-Strategy. In order to take this position, you should first point ou that a person S is not able to distingsuish between p and q, then - secondly - you should claim that in order to know that p a person S must have excluded the relevant alternatives that p. ...and be unhappy.

Third possibility: Just be a happy Boghossian- strategist. Point out, that "water" is an indexical and that it corresponds with our intuitions that we have to undertake empirical investigations in order to know what we think.
Happy or unhappy?

:-) enwe

I know that there are several options for tackling the externalism problem. But as I have no sympathies with the externalist theories that traditionally generate the problem, my hope is that I can simply avoid having to choose.

Anyway, the point I wanted to make is not that the familiar externalist problems arise if one accepts causal conditions in the individuation of mental states. If my reasoning was correct, the externalism we get is unfamiliar and bizarre: the content of my water-thoughts depends not only on what occupies the water-role in my surroundings (that's familiar) but just as much on what happens in any far away galaxy. Likewise for the phenomenal quality of my pain experiences. That should sound wrong even to a die-hard externalist. So there must be a fault in my reasoning. The question is where.

sorry, mea culpa.

you're either conflating different forms of possibility or you're begging the question. if we're restricting the domain of possible worlds to those that are nomologically possible, then there will be no possible world in which the physical laws are drastically different (or perhaps even different at all).

if we don't place such a restriction on W, then scientific essentialists will argue that modal intuitions about mental properties differing across possible worlds, even when realized by identical physical properties, are mistaken or at least mistakenly reported.

if being in pain=having firing c-fibers, then necessarily being in pain=having firing c-fibers. it doesn't matter what possible world we're talking about and how drastically different the physical laws are. we will always be in the same mental state as our possible world counterparts. of course, having firing c-fibers is a poor example of a physical property which might be essential to being in pain, but *surely* there's a first-order physical property common to all pain realizers that's essential to being in pain. if so, then it's not possible that i could be in pain in the actual world while my counterpart not be in a physical world with different physical laws.

have you read John Hawthorne's paper on why Humeans are out of their minds? Its kinda related to what you worry about here..

Rich: thanks, I'll have a look at the Hawthorne paper. The problem is more obvious for Humeans about laws, but I think it applies to other accounts as well. E.g. there are many different necessitation relations between universals compatible with what goes on in our galaxy, and I assume there is no direct way to find out the actual necessitation relations.

muxol: yes, if the individuation of mental states is a matter of intrinsic physical nature the problem doesn't arise (even though a similar one probably arises if we consider world 2 as actual). However, I find that very implausible. It implies that it's impossible for beings with a very non-human body to have desires, beliefs, perceptions, or pains.

wo, is it really all that implausible though? what we consider as desires, beliefs, and pains are probably terms ranging over a broad category of species-specific mental states realized by very similar neurophysiological systems.

what constitutes a very non-human neurophysiological system? surely even a rat and human have very comparable neurophysiological systems.

Yes, John Hawthorne's argument is somewhat similar to my worries. But I took it for granted that if certain causal roles are essential to a mental state then being in that state is not an intrinsic property. Apparently, on Hawthorne's view the causal relations that enter into the role all relate the mental state to other intrinsic states of the subject, but never to external things. This is a rather odd position, as Brian Weatherson has pointed out.

So I took it for granted that mental states are not completely intrinsic to their subjects. I don't find that all too implausible. What I find implausible is that my mental states are not intrinsic to our galaxy.

In his reply to Hawthorne, Brian incidentally suggests an answer to my worry: Causal properties might not depend on universal laws but only on "local laws", lawlike regularities obtaining in the relevant region of the universe. I guess the solution must look somehow like this, but I wonder how that solution looks like in more detail: Surely the regions that matter for causal properties can't be arbitrarily small (or large, for that matter).